Determinants of Small Business Performance: A Meta-Analysis

Determinants of Small Business Performance: A Meta-Analysis Osita C. Nwachukwu Fairleigh Dickinson University Braimoh Oseghale Fairleigh Dickinson University Please address all correspondence to the first author. Abstract The links between small business performance and its determinants were meta-analyzed in this study. A search of the literature uncovered 27 articles that yielded 50 correlations. We found a significant relationship between the predictor variables and ROA, ROS, and ROI. However, the size of the effect was quite small. The importance of small business to the economic health of the nation is generally acknowledged by academicians and practitioners. According to the Small Business Administration (2008), 99.7 percent of all businesses in the United States are small. They also employ 50.6 percent of the non-farm private sector workforce. In spite of the impressive numbers, research on determinants of small business performance has been scanty. A number of researchers have, as Rutherford and Oswald (2000) point out, examined the impact of individual characteristics, firm characteristics, and environmental characteristics on small business performance. The results of previous empirical studies on determinants of small business performance have been inconclusive. A major purpose of the present study is to cumulate the findings of empirical research on determinants of small business performance. A second purpose is to use a larger sample of studies to estimate population values for the relationships between small business performance and its antecedents. Previous Research on Determinants of Small Business Performance Inquiries & Perspectives 65 Volume 3 Number 1 October 2010

As Cragg and King (1988) and Rutherford and Oswald (2000) observe, previous research on determinants of small business performance fell into three categories: individual characteristics, firm characteristics, and environmental characteristics. Individual Characteristics of the Firm Studies that fell under this category have examined the relationship between individual characteristics and performance such as: age, education, managerial experience, industry experience, leadership practices, race, CEO personality, and gender (Foley, 1985; Begley & Boyd, 1986; Lussier, 1995; Steiner & Solem, 1988; Miller and Toulouse, 1986; Fasci & Valdez, 1998; Frith, 1998; Ozcelik et al., 2008). Characteristics of the Firm Studies that fell under this category have examined firm characteristics such as strategy/planning, structure, competitive orientation, top management team, culture, organizational growth, family control, operations management, and stage of development (Robinson, et al., 1984; Riggs & Bracker, 1986; Miller & Toulouse, 1986; Bracker and Pearson, 1986; Gable & Topol, 1987; Bracker, et al., 1988; Weinzimmer, 1997; Stoica & Schindelmitte, 1999; Lerner & Almor, 2002; Pleshko, 2006; Megicks, 2007; Danes, et al., 2008; Oswald, et al., 2009). Characteristics of the Environment Studies that fell under this category have examined contacts with customers, suppliers, competitors, regulatory organizations, consultants, creditors, stockholders, and financial institutions. Other aspects of the environment include perceived uncertainty in the industry environment (Dollinger, 1985; Shrader, et al., 1989; Sawyerr, 2003). Method Sample We conducted an extensive search in order to identify studies examining the relationship between small business performance and its antecedents. First, we used computer-aided keyword searches of ABI Inform, Business Service Premier, and JSTOR using keywords small business performance and determinants of small business performance. Second, we manually researched key journals in various business disciplines (e.g. Academy of Management Journal, Administrative Science Quarterly, American Journal of Small Business, Entrepreneurship Theory and Practice, Journal of Applied Psychology, Journal of Business and Entrepreneurship, Journal of Business Strategies, Journal of Business Venturing, Journal of Management, Journal of Management Studies, Journal of Managerial Issues, Journal of Managerial Psychology, Journal of Small Business Management, Journal of Small Business Strategy, Long-Range Planning, Management Decision, Management Science, and Small Business Forum. Third, we also employed what has been termed a snowballing procedure by Davis and Rothstein (2006). A snowballing procedure is scanning of the references included in the relevant studies to identify other relevant studies. To be included in our meta-analysis, studies had to report a Pearson product-moment correlation, an f- Inquiries & Perspectives 66 Volume 3 Number 1 October 2010

statistic, t-statistic, or chi-squares with their corresponding degrees of freedom. Overall, our search produced 27 studies, with 50 effect sizes, and a total sample size of 15,543. The sample size is derived from adding the number of companies on which each of the 27 studies relied. Study Table 1 Studies Included in Meta-Analysis Sample Independent Size Variable Dependent Variable Robinson, Pearce, Vozikis, and Mescon (1984) 51 Stage of Development Sales, ROS # of Employees Ackelsberg and Arlow (1985) 135 Planning Growth in Sales Profit Index Dollinger (1985) 82 Environmental Contact Orpen (1985) 52 Long-Range Planning Sales, Net Income Profit Index ROA Bracker and Pearson (1986) 188 Planning Dollinger and Kolchin (1986) 81 Boundary Spanning Activity Miller and Toulouse (1986) 97 Strategy, Structure and CEO Personality Riggs and Bracker (1986) 183 Operations Management Profit Index Profit ROI Sales Growth Gable and Topol (1987) 179 Planning Sales Profits Bracker, Keats, and Pearson (1988) 73 Planning Net Income Cragg and King (1988) 179 Organizational Characteristics Covin and Slevin (1989) 161 Environmental Hostility, Structure, Strategic Posture, Competitive Tactics Profit Sales Profit Index Inquiries & Perspectives 67 Volume 3 Number 1 October 2010

Table 1 (Continued) Study Sample Size Independent Variable Dependent Variable Shrader, Mulford, and Blackburn (1989) 97 Strategic and Operational Planning, and Environmental Uncertainty Sales Net Income Kalleberg and Leicht (1991) 312 Gender Net Income Weinzimmer (1997) 74 Top Management Team Variables Fasci and Valdez (1998) 604 Male and Female Owned Frith (1998) 197 Market Orientation, Minority, and Woman-Owned Profit Index ROS Kean, Gaskill, Leistritz, Jasper, Bastow-Shoop, Jolly, and Sternquist (1998) 456 Community Characteristics, Business Enviroment, and Competitive Strategies ROS Stoica and Schindehutte (1999) 242 Adaptability Profit Index Lerner and Almor (2002) 220 Strategic Capabilities and Management Styles Sawyer, McGee, and Peterson (2003) 153 Perceived Environmental Uncertainty Wang and Ang (2004) 40 Environment, Resource-based Capabilities, Strategy, and Venture Capital backed Firm s Involvement Pleshko (2007) 125 Strategic Orientation, and Organizational Structure Sales Net Income # of Employees Net Income ROA Sales # of Employees Sales Profits Inquiries & Perspectives 68 Volume 3 Number 1 October 2010

Table 1 (Continued) Study Sample Size Independent Variables Dependent Variables Megicks (2007) 305 Levels of Strategy ROI Danes, Teik-Cheok loy, and Stafford (2008) 572 Planning Ozcelik, Langton, and Aldrich (2008) 229 Leadership Practices Oswald, Muse, and Rutherford (2009) 2631 Percent of Family Control Company Performance Revenue Capital Structure Meta-analyses were conducted using Hunter and Schmidt (1990) procedures. It is a technique that allows one to aggregate correlation coefficients across empirical studies to derive unbiased estimates of population relationships by correcting for the presence of statistical artifacts. Hunter and Schmidt (1990) suggest that the best estimate of the size of the correlation between two variables is the weighted average in which each correlation is weighted by the number of subjects in that study. A number of studies included in our sample contain multiple measurements of predictor and criterion variables. As Volckner and Hofmann (2007) observe, studies with multiple effect sizes may have a greater impact on the results of the metaanalysis than studies that only contribute one effect size. Bijmolt and Pieters (2001) suggest two general approaches for dealing with multiple measurements: The first approach is to represent each study by a single value, such as an average effect size (Hunter & Schmidt,1990). A second approach is the complete set approach. Under this approach, the values of all measurements within the studies are incorporated and treated as independent (weighted) replications (Kirca et al., 2005; Tellis, 1988; Volckner & Hofmann, 2007). As Bijmolt and Pieters (2001), Volckner and Hofmann (2007) observe, this single value approach results in a serious loss of information. We chose to employ the complete set approach because Bijmolt and Pieters (2001) demonstrated the superiority of this approach in a Monte Carlo study and a reanalysis of a published marketing data set. Recognizing that studies with many measurements may have a greater effect than studies with fewer or single measurements on the results of our metaanalysis, we chose to adopt the Volckner and Hofmann (2007) approach of weighting the effect sizes by the inverse of the number of multiple measures in the study. Results Since studies examining the impact of various predictor variables on small business performance have generally fallen into three groups: individual characteristics, firm characteristics, and environmental characteristics (Rutherford & Oswald, Inquiries & Perspectives 69 Volume 3 Number 1 October 2010

2000), all 27 studies were placed in those three categories. Six studies were included in more than one category. In addition, profitability index or net income. Seven separate meta-analyses were performed in this study, one involving all 27 studies and since the effect size seems to vary from one one each on the three groups of performance indicator to another (Boyd, performance indicators and the three 1991; Schwenk & Shrader, 1993) metaanalyses categories (individual, firm, and were performed separately for environment) for classifying small business three sets of performance measures: ROA, performance studies. All seven metaanalyses ROI, and ROS; sales or revenue growth; and are shown in Table 2. Table 2 Meta Analysis Results Category N R σ 2 r σ 2 e σ 2 p r/σp All Studies 15,543.05.005.003.002 1 ROS, ROA, ROI 1,311.08.007.005.002 2 Megicks 305.22 Sawyerr et al. 153.11 Robinson et al. 51.13 Orpen 52.06 Miller & Toulouse 97.02 Frith 197.02 Kean et al. 456.02 Sales/Revenue Growth 7,968.04.006.003.003 0.8 Shrader et al. 97.05 Sawyerr et al. 153.08 Cragg & King 179.02 Robinson et al. 51.12 Dollinger 82.05 Orpen 52.08 Riggs & Bracker 183.19 Miller & Toulouse 97.03 Bracker et al. 73.21 Bracker & Pearson 188.22 Ackelsberg & Arlow 135.14 Gable & Topol 179.04 Ozcelik et al. 229.06 Wang & Ang 40.13 Frith 197.09 Weinzimmer 74.23 Oswald et al. 2,631 -.001 Oswald et al. 2,631 -.003 Danes et al. 572.17 Pleshko 125.28 Inquiries & Perspectives 70 Volume 3 Number 1 October 2010

Table 2 (Continued) Category N R σ 2 r σ 2 e σ 2 p r/σp Net Income/Profit Index 2,822.08.004.006 -.002 - Shrader et al. 97.02 Sawyerr et al. 153.05 Cragg & King 179 -.02 Dollinger 82.05 Dollinger 82.02 Miller & Toulouse 97.03 Dollinger & Kolchin 81.22 Bracker et al. 73.10 Ackelsberg & Arlow 135.09 Gable & Topol 179.02 Stoica & Schindehutte 242.09 Fasci & Valdez 604.15 Kalleberg & Leicht 312.03 Covin & Slevin 161.06 Pleshko 125.19 Lerner & Almor 220.04 Individual Characteristics 11,184.02.003.001.002.44 Kalleberg & Leicht 312.03 Ozcelik et al. 229.02 Ozcelik et al. 229.06 Frith 197.02 Frith 197.09 Lerner & Almor 220.08 Lerner & Almor 220.04 Lerner & Almor 220.04 Danes et al. 572.17 Oswald et al. 2,631 -.001 Oswald et al. 2,631 -.003 Oswald et al. 2,631.001 Fasci & Valdez 604.15 Miller & Toulouse 97.03 Miller & Toulouse 97.03 Miller & Toulouse 97.02 Inquiries & Perspectives 71 Volume 3 Number 1 October 2010

Table 2 (Continued) Category N R σ 2 r σ 2 e σ 2 p r/σp Firm Characteristics 3,889.11.007.007 0 - Pleshko 125.28 Pleshko 125.19 Cragg & King 179 -.02 Cragg & King 179.02 Wang & Ang 40.16 Wang & Ang 40.13 Kalleberg & Leicht 312.03 Bracker et al. 73.21 Bracker et al. 73.10 Megicks 305.22 Danes et al. 572.17 Shrader et al. 97.05 Shrader et al. 97.02 Weinzimmer 74.23 Bracker & Pearson 188.22 Gable & Topol 179.04 Gable &Topol 179.02 Riggs & Bracker 183.19 Miller & Toulouse 97.03 Miller & Toulouse 97.03 Miller & Toulouse 97.02 Robinson et al. 51.12 Robinson et al. 51.13 Robinson et al. 51.07 Robinson et al. 51.03 Orpen 52.08 Orpen 52.06 Ackelsberg & Arlow 135.14 Ackelsberg & Arlow 135.09 Environmental Characteristics 1873.07.002.006 -.004 - Frith 197.02 Frith 197.09 Wang & Ang 40.16 Wang & Ang 40.13 Sawyerr et al. 153.05 Sawyerr et al. 153.08 Sawyerr et al. 153.11 Covin & Slevin 161.06 Kean et al. 456.02 Stoica & Schindehutte 242.09 Dollinger & Kolchin 81.22 Inquiries & Perspectives 72 Volume 3 Number 1 October 2010

As Table 2 shows, the cumulated effect size across all 50 correlations produced an r of.05. This effect size was based on an overall sample size of 15,543 firms. The comparison of the overall corrected standard deviation of.05 to the mean of.05 is only 1.0 standard deviation above zero. This is a borderline result. The probability of a zero or below zero correlation, however, cannot be ruled out. From a qualitative perspective, the population correlation is positive for all the studies. The effect size for ROS, ROA, and ROI is.08, and is based on a sample size of 1,311. A comparison of the corrected standard deviation for these group of studies of.04 to the mean of.08 is two standard deviations above zero. So the probability of a zero or below zero correlation with this group is highly unlikely. Even though the effect size of.8 is significant, it still amounts to.6 percent of the variation in small business ROS, ROA, and ROI. In other words, the average determinant only accounts for.6 percent of the population variation. For the sales or revenue growth group, there were 20 correlations ranging from -.003 to.28. The sample size was 7,968, and the effect size was.04. A comparison of the corrected standard deviation of.05 to the mean of.04 is.8 standard deviation above zero. So the probability of a zero or below zero correlation cannot be ruled out. The net income or profitability index group had an effect size of.08, and is based on a sample size of 2,822. An r/6p comparison was not meaningful in this group since σ 2 p had a -.002 value. Sampling error could not be ruled out. The effect size for the individual characteristics group is.02 and is based on a sample size of 11,184. A comparison of the corrected standard deviation for this group of studies of.045 to the mean of.02 is only.44. Less than one standard deviation above zero. So the probability of a zero or below zero correlation cannot be ruled out. The effect size for firm characteristics is.11, and is based on a sample size of 3,889. An r/σp comparison was not meaningful in this group since σp had a zero value. Sampling error could not be ruled out for this group. The effect size for the environmental characteristics group is.07, and is based on a sample size of 1,873. An r/σp comparison was not meaningful in this group also, since σ 2 p had a -.004 value. Again, sampling error could not be ruled out. Discussion As Boyd (1991) observes, the presence of measurement error will consistently lower the estimate of a correlation coefficient or effect size. We did not correct the observed effect sizes in this study for measurement error because most of the studies did not report reliabilities for the predictor and dependent variables. If we had been able to correct for measurement error (or attenuation) some of the borderline effect sizes reported here may have reached significance. Future Research In light of the lack of reporting of reliability coefficients in the studies in our meta-analysis, one obvious remedy would be for future researchers to report the reliabilities of the measures in their studies. A second suggestion would be to use multiple indicators to measure variables of Inquiries & Perspectives 73 Volume 3 Number 1 October 2010

interest, especially small business performance (Bagozzi & Phillips, 1982; Keats, 1983; Boyd, 1991). In light of the very low average effect size of.08 observed in our ROA, ROS, and ROI group meta-analysis, another suggestion for future researchers would be to include more predictors in their studies. Ketchen et al. (1997), in their metaanalysis of configuration and performance relationship, found that studies using longitudinal designs reported larger effect sizes. We would suggest more longitudinal studies for this reason, and also in order to demonstrate causality. Our hope is that once these suggestions are incorporated, a future meta-analysis will be able to observe stronger effect sizes. References Ackelsberg, R. & Arlow, P. (1985). Small businesses do plan and it pays off. Long Range Planning, 18, 61-67. Bagozzi, R. P. & Phillips, L. H. (1982). Representing and testing organizational theories: A holistic construal. Administrative Science Quarterly, 17, 459-489. Begley, T. M. & Boyd, D. P. (1986). Executive and corporate correlates of financial performance in smaller firms. Journal of Small Business Management, 26, 8-15. Boyd, B. K. (1991). Strategic planning and financial performance: A meta-analytic review. Journal of Management Studies, 28, 353-375. Bracker, J. S. & Pearson, J. N. (1986). Planning and financial performance of small, native firms. Strategic Management Journal, 7, 503-522. Bracker, J. S., Keats, B. W., & Pearson, J. N. (1988). Planning and financial performance among small firms in a growth industry. Strategic Management Journal, 9, 591-603. Bijmolt, T. H. A. & Pieters, R. G. M. (2001). Metaanalysis in marketing when studies contain multiple measurements. Marketing Letters, 12, 157-169. Covin, J. C. & Slevin, D. P. (1989). Strategic management of small firms in hostile and benign environments. Strategic Management Journal, 10, 75-87. Cragg, P. B. and King, M. (1988). Organizational characteristics and small firms performance revisited. Entrepreneurship Theory and Practice, 13, 49-64. Danes, S. M., Teik-Cheok Loy, J. & Stafford, K. (2008). Business planning practices of family-owned firms within a quality framework. Journal of Small Business Management, 46, 395-421. Davis, A. L. and Rothstein, H. R. (2006). The effects of the perceived behavioral integrity of managers on employee attitudes: A metaanalysis. Journal of Business Ethics, 67, 407-419. Dollinger, M. J. (1985). Environmental contacts and financial performance of the small firm. Journal of Small Business Management, 23, 24-30. Dollinger, M. J. & Kolchin, M. G. (1986). Purchasing and the small firm. Journal of Small Business, 10, 33-45. Fasci, M. A. & Valdez, J. (1998). A performance contrast of male- and female-owned small accounting practices. Journal of Small Business Management, 36, 1-7. Foley, P. (1985). What makes a small business successful? Occasional paper 85/41, Sheffield Centre for Environmental Research. Frith, J. R. (1998). Market orientation versus performance in minority and womanowned small business. Proceedings of the Academy of Marketing Studies, 3, 6-19. Gable, M. & Topel, M. T. (1987). Planning practices of small-scale retailers. American Journal of Small Business, 12, 19-32. Hunter, J. E. & Schmidt, F. L. (1990). Methods of meta-analysis: Correcting error and bias in research findings. Newbury Park, CA: Sage Publications. Kalleberg, A. L, & Leicht, K. T. (1991). Gender and organizational performance: Determinants of small business survival and success. Academy of Management Journal, 34, 136-161. Kean, R., Gaskill, L., Leistritz, L., Jasper, C. Bastow- Shoop, H., Jolly, L., & Sternquist, B. (1998). Effects of community characteristics, business environment, and competitive strategies on rural retail business Inquiries & Perspectives 74 Volume 3 Number 1 October 2010

performance. Journal of Small Business Management, 36, 45-57. Keats, B.N. (1983). On the measurement of strategic performance. Paper presented at the Strategic Management Society Annual Conference, Paris. Ketchen, D. J., Combs, J. G., Russell, C. J., Shook, C., Dean, M. A., Runge, J., Lohrke, F. T.,Naumann, S. E., Haptonstahl, D. E., Baker, R. Beckstron, B. A., Handler, C. Honig, H., & Lamoureaux, S. (1997). Organizational configurations and performance: A metaanalysis. Academy of Management Journal, 40, 223-240. Kirca, A. H., Jayachandran, S. & Bearden, W. O. (2005). Market orientation: A meta-analytic review and assessment of its antecedents and impact on performance. Journal of Marketing, 69, 24-41. Lerner, M. & Almor, T. (2002). Relationships among strategic capabilities and the performance of women-owned small ventures. Journal of Small Business Management, 40, 109-125. Lussier, R. N. (1995). A nonfinancial business success versus failure prediction. Journal of Small Business Management, 33, 1-11. Megicks, P. (2007). Levels of strategy and performance in UK small retail businesses. Management Decision, 45, 484-502. Miller, D. & Toulouse, J. M. (1986). Strategy, structure, CEO personality and performance in small firms. American Journal of Small Business, 10, 47-62. Orpen, C. (1985). The effects of long-range planning on small business performance: A further examination. Journal of Small Business Management, 23, 16-23. Oswald, S. L., Muse, L. A. & Rutherford, M. W. (2009). The influence of large stake family control on performance: Is it agency or entrenchment? Journal of Small Business Management, 47, 116-135. Ozcelik, H., Langton, N. & Aldrich, H. (2008). Doing well and doing good: The relationship between leadership practices that facilitate a possible emotional climate and organizational performance. Journal of Managerial Psychology, 23, 186-203. Pleshko, L. P. (2007). Strategic orientation, organizational structure, and the associated effects on performance. Journal of Financial Services Marketing, 12, 53-64. Riggs, W. E. & Bracker, J. S. (1986). Operations management and financial performance. American Journal of Small Business, 10, 17-23. Robinson, R. B. Pearce, J. A., Vozikis, G. & Mascon, T. (1984). The relationship between stage of development and small firm planning and performance. Journal of Small Business Management, 22, 45-52. Rutherford, M. W. & Oswald, S. L. (2000). Antecedents of small business performance. New England Journal of Entrepreneurship, 3, 21-33. Sawyerr, O., McGee, J. & Peterson, M. (2003). Perceived uncertainty and firm performance in SMEs: The role of personal networking activities. International Small Business Journal, 21, 269-288. Schwenk, C. R. & Shrader, C. B. (1993). Effects of formal strategic planning on financial performance in small firms: A metaanalysis. Entrepreneurship Theory and Practice, 17, 53-64. Shrader, C. B. Mulford, C. L., & Blackburn, V. L. (1989). Strategic and operational planning, uncertainty, and performance in small firms. Journal of Small Business Management, 27, 45-60. Steiner, M. P. & Solem, O. (1988). Factors for success in small manufacturing firms. Journal of Small Business Management, 36, 51-56. Stoica, M. & Schindehutte, M. (1999). Understanding adaptation in small firms: Links to culture and performance. Journal of Developmental Entrepreneurship, 4, 1-18. Tellis, G. J. (1988). The price elasticity of selective demand: A meta-analysis of econometric models of sales. Journal of Marketing Research, 25, 331-341. United States Small Business Administration. (2008). Tables of size standards, URL (consulted January 2009): http://www.sba.gov/size/indextableofsize.h tml. Volckner, F. & Hofmann, J. (2007). The priceperceived quality relationship: A metaanalytic review and assessment of its determinants. Marketing Letters, 18, 181-196. Wang, C. K. & Ang, B. (2004). Determinants of venture performance in Singapore. Journal of Small Business Management, 42, 347-363. Inquiries & Perspectives 75 Volume 3 Number 1 October 2010

Weinzimmer, L. G. (1997).Top management team correlates of organizational growth in a small business context: A comparative study. Journal of Small Business Management, 35, 1-9. Osita C. Nwachukwu Associate Professor of Management Fairleigh Dickinson University Madison, NJ 07940 Phone: 973-443-8860 E-Mail: osita@fdu.edu Braimoh Oseghale Associate Professor of International Business Fairleigh Dickinson University Teaneck, NJ 07666 Phone: 201-692-7267 E-Mail: oseghale@fdu.edu Inquiries & Perspectives 76 Volume 3 Number 1 October 2010